Introduction

A verification study has been carried out to validate the Bladed turbulence generation code against theory. Bladed generates three-dimensional turbulent wind fields containing time histories of wind speed variations over a user-defined grid which spans in lateral and vertical dimensions. Two following turbulence models in Bladed are evaluated:

• Kaimal turbulence model, and
• Mann turbulence model.

In the case of the Kaimal turbulence model, the Veers method is employed to generate wind files. Under this method, each turbulence component is defined by a power spectral density function and coherence function. The power spectral density function defines the frequency content of the time history at each grid point in the wind file. The coherence function describes the correlation of the turbulence time histories between grid points in the wind file, which are separated laterally and vertically as a function of frequency. As the coherence is a function of separation, points that are close together will be more highly correlated and therefore have greater coherence than those which are located further away from each other.

The Mann turbulence model model follows a different approach to generate the wind files. This methodology involves determining the spectral density for any three-dimensional wavenumber vector, followed by the simultaneous generation of all three turbulence components through the summation of such wavenumber vectors, each assigned with appropriate amplitude and random phase. Notably, this process necessitates the use of a three-dimensional fast Fourier transform (FFT) to ensure computational efficiency and accuracy.

The verification for the turbulent wind generation in Bladed was carried out by comparing the spectra and coherence against the theoretical quantities. This is done for each investigated turbulence model according to the data processing method which explains the general settings for the wind file generation and the calculations of the statistical quantities. The verification for the Kaimal turbulence model is presented in terms of the spectra and coherence for two different grid spacings and for different separation distances between grid samples. Similarly, the Mann turbulence model is also verified based on the spectra and coherence. However, in addition to that, descriptions of the FFT points significance for generating the turbulent wind data using the Mann model are also provided. This is in conjunction with the best practice for determining the number of grid for the Mann model to enable an accurate representation of the turbulent flow field.

Last updated 17-07-2024